Spread the love

Post Views: 49

In the great software race of 2025, it seems artificial intelligence isn’t just coming for your job – it’s interviewing itself and rewriting the onboarding process. According to a compelling new study by AI-powered low-code platform OutSystems, an eye-watering 93% of software executives are either developing or planning to develop their custom AI agents. Yes, you read that correctly – ninety-three per cent.

The report, colourfully titled Navigating Agentic AI & Generative AI in Software Development: Human-Agent Collaboration is Here, was compiled in partnership with CIO Dive and KPMG. It pulls no punches in revealing the extent to which agentic AI is not only infiltrating development workflows but swiftly reshaping the DNA of software teams from Boston to Bangalore.

And if you thought AI was all talk and no torque, think again. This isn’t speculation – it’s happening. And fast.

The Rise of the Machines (with Manners)

As the report notes, today’s IT leaders are walking a tightrope – expected to deliver business value, drive digital transformation and juggle legacy systems, all while their teams are maxed out and their tools resemble a tangled ball of yarn. Enter agentic AI: digital entities built not only to assist but to act semi-autonomously, pulling data across siloes, refining code, monitoring processes, and, if you believe the boldest predictions, occasionally making the tea.

“Software development is undergoing a seismic transformation,” declared OutSystems CEO Woodson Martin, who, if he had a dollar for every time the word ‘agent’ was used in this report, could probably buy ChatGPT a new server farm. “In the very near future, AI agents will function like specialised digital departments – constantly learning, optimising, and refining software to align with business priorities. Developers and executives alike will be liberated to focus on more creative, strategic initiatives.”

So the robots won’t take your job – they’ll take the bits you don’t like. Or that’s the pitch, at least.

Beyond the Buzzwords: What the Data Says

This wasn’t some throwaway LinkedIn poll. The OutSystems study surveyed 550 software executives from across the globe – including the US, UK, Australia, Germany, France, Japan, India and Canada – representing industries from manufacturing to finance. The consensus? AI isn’t just functional. It’s indispensable.

The report’s headline figures are impressive:

More than 66% of respondents reported enhanced developer productivity.
62% saw improved scalability in their development pipelines.
60% said testing and QA got a leg-up thanks to AI involvement.
And perhaps most importantly, bugs were down, quality was up.

This signals a step change in software development – one that doesn’t just speed things up but redefines what developers do.

New Skills, New Roles, New Frontier

You can’t unleash a digital revolution without shifting a few desks. And that’s precisely what’s happening.

OutSystems found that 69% of executives anticipate the emergence of new specialist roles—think prompt engineers, agent architects, and those overseeing AI orchestration. Meanwhile, 63% foresee a serious need for upskilling and reskilling existing staff to keep pace with their newly minted silicon colleagues.

Michael Harper, Managing Director at KPMG LLP, summed it up neatly: “A lot of organisations began with pilots a year or two ago, but now they’re seeing real gains in code generation and testing. That’s fuelling confidence – and investment.”

From Help Desk to Humanless

Customer service is emerging as agentic AI’s favourite playground. 49% of respondents plan to deploy AI agents to handle queries and streamline support autonomously. But the buck doesn’t stop there. Other areas in the crosshairs include:

Product development (38%)
Sales and marketing (32%)
Supply chain management (28%)
Human resources (24%)
Finance and accounting (23%)

In other words, anywhere a human’s patience has worn thin, an AI agent with a polite tone and zero lunch breaks can thrive.

The Elephant in the Server Room: Risk and Governance

Of course, not all that glitters is algorithmic gold. OutSystems does not mince words regarding the shadow side of AI adoption.

64% of executives flagged governance, security and compliance as pressing concerns. Transparency and reliability of AI decision-making also emerged as major sticking points.

Even more alarming? 44% are already experiencing “AI sprawl” – a messy proliferation of overlapping tools and frameworks leading to fragmentation and technical debt.

The message is clear: the agentic AI gold rush is well underway, but organisations risk burying themselves under their digital ambitions without a blueprint for responsible scaling.

Human + Machine: The Future’s Favourite Formula

OutSystems’ findings echo a broader industry consensus: the most successful AI strategies will be collaborative, not competitive. By offloading repetitive tasks to AI, human developers can innovate, strategise, and deliver richer, more complex user experiences.

As AI agents mature, they’ll act more like trusted colleagues than robotic assistants—identifying opportunities, triggering improvements, and making businesses smarter from the inside out.